Evaluating best model (dauntless-sweep-2) with both NMS and MOB bounding box aggregation and both evaluation schemes: standard PASCAL VOC2012 and our SAR-APD (HERIDAL) evaluation

Below are plotted precision, recall and mAP for both train and test data for different score threshold values (calibration parameter of the inference model). Score threshold values above 0.1 seem to give better trade off between precision and recall for both data sets.

testing evaluation script